The linkages between drugs and violence are poorly understood. Both substance use and serious interpersonal violence are first manifested in the mid-adolescent years, but generally start to decrease by the early 20s. However, violent behavior occurs at substantially younger ages than drug use, and rates decline earlier and more sharply. The overall goal of this study is to increase our understanding of the drug relatedness of violence in the sharply. The overall goal of this study is to increase our understanding of the drug relatedness of violence in the critical 14-17 year old age range. We will conduct a multi-group (students, institutionalized juveniles, and dropouts) cross-national (US and Canada) study of the experiences and attitudes of 2200 youth in two large metropolitan areas-Philadelphia and Toronto. The study will address these research questions: 1) do drugs contribute to a significant amount of youth violence?, 2) what are the patterns and correlates of drug-related violence among youth? , 3) how does Goldstein's tri-partite model of the drugs-violence nexus apply to adolescents?, 4) how do the structural features of neighborhoods and other policy and environmental factors impact on drug-related violence? , and 5) can commonalities and differences identified in the two cities explain the relationships? By defining interpersonal violence as overt behavior directed by one person against another, intended to inflict physical pain or injury, this study will encompass a broad range of violent acts from minimal to extreme seriousness. As well, by incorporating the threat of violence, or verbal aggression, this study acknowledges the public health concern with the harmful psychological consequences of violence. All drugs with psychoactive effects, including alcohol, will be investigated. The concept of social capital, applied in recent innovative empirical studies of adolescent deviance, provides the broad theoretical framework in which to assess the relative importance of individual, family, environmental, and neighborhood characteristics on the outcome measures of drug-related violence. In addition, these data will allow consideration of crucial policy questions about what forms of personal and social interventions might be effective in preventing or reducing drug-related violence among youth.